Introduction
In contentious maritime disputes, the International Court of Justice uses a ‘three-stage’ approach to delineate maritime boundaries. This comprises (a) the drawing of a provisional equidistance line; (b) a consideration of ‘relevant circumstances’ which call for the adjustment of this line to achieve an equitable result; and (c) the establishment of a disproportionality test to ensure that the line does not lead to an inequitable result.[1] At the second stage of the test, the Court has generally adopted a narrow focus based on geographical factors, such as distance, coordinates, and coastal measurements, often leading to the exclusion of non-geographical factors like security and the interests of local populations. This is highlighted in the recent Somalia v Kenya case: while Kenya invoked a range of non-geographical factors, they were superficially considered and largely glossed over. This article aims to critique the Court’s limited engagement with non-geographical factors in maritime delimitation cases.
This essay will first look at the importance of maritime boundaries. It will then examine the problems in the Court’s current approach. Additionally, it will be posited that the Court’s reliance on geographical factors stems from a misunderstanding of its judicial role, as it adheres rigidly to a formalist interpretation of sovereignty and incorrectly equates geophysical boundaries with border stability. Finally, based on the aforementioned arguments, it will be contended that the Court should consider non-geographical factors with equal scrutiny at the same stage of the test.
Why do maritime boundaries matter?
Maritime zones are drawn according to the coastal state’s territory, where the breadth of maritime zones is measured from baselines drawn along the coast.[2] When states have overlapping claims to maritime zones, maritime boundaries are drawn to apportion these zones to competing states. Thus, maritime zones are linked closely to the expansion of finite territorial space and the consolidation of sovereign rights.[3] For instance, exclusive economic zones confer the sovereign rights to conserve, explore, exploit, and manage natural resources, and contiguous zones empower states to exercise the necessary control to avert the infringement of its fiscal, immigration, customs, or sanitary laws.[4] Per Hohfeld’s theory of rights, maritime boundaries are legally significant because they create both liberties and privileges to act, such as in mining natural resources, and claim-rights, which refer to the rights to have other states avert the infringement of domestic laws.[5] This theory is analytically useful because it shows that rights arising from maritime boundaries extend far beyond mere rights to exercise one’s sovereignty, encompassing rights to limit others’ sovereignty under one’s jurisdiction.
The consideration of ‘relevant circumstances’ in Somalia v Kenya
Somalia initiated proceedings against Kenya in 2014 for the delimitation of their maritime boundary in the Indian Ocean, asking the Court to determine a provisional equidistance line without further adjustment.[6] Kenya contended that there was a pre-existing maritime boundary drawn according to a parallel of latitude,[7] and that modification was needed even if the provisional line were adopted to achieve an equitable result. [8]
The Court employed the three-stage test, first drawing a provisional equidistance line.[9] At the second stage of the test, Kenya invoked multiple non-geographical ‘relevant circumstances’ to support its claim that the provisional equidistant line should be adjusted.[10] In particular, Kenya cited the ‘vital security interests’ of both the parties and international community. Namely, Kenya contended that it incurred significant costs from regularly patrolling its waters to regulate weapon smuggling and piracy, perpetrated by Al-Shabaab, a terrorist militant group based in Somalia.[11] Kenya also argued that the parties’ long-standing and consistent conduct in relation to oil concessions, naval patrols, fishing, and other activities, reflected the existence of a de facto maritime boundary.[12] For instance, while Somalia attempted to dispute this by arguing that Kenya’s surveys and maps of fishing activities lacked evidentiary value,[13] and that Somalia was unaware of Kenya’s fishing regulations,[14] Kenya counterargued that Somalia’s own official practice demonstrated its acceptance of Kenya’s fishing zones.[15] Lastly, Kenya referred to the ‘devastating repercussions’ for the livelihoods and economic well-being of its fishing communities.[16]
Despite stating that the ‘relevant circumstances’ are ‘mostly geographical in nature’, but still emphasising that there is ‘no closed list of relevant circumstances’,[17] the Court ultimately declined to accept any of the non-geographical factors invoked by Kenya. This gap between the Court’s theoretical acceptance of non-geographical factors and its eventual use of purely geographical factors was insufficiently reasoned. In addressing Kenya’s security concerns, the Court merely stated that the security situation was ‘not of a permanent nature’, whereas maritime boundaries were ‘aimed at providing permanency and stability’.[18] This distinction between permanence and impermanence inherently excludes non-geographical factors because they are seen as mutable and ever-changing. This reasoning is self-evident, with the Court essentially arguing that non-geographical factors should not be counted due to their essential nature as non-geographical factors, while geographical factors should be counted because of their essential nature as geographical factors. Additionally, in addressing Kenya’s argument of the impact on its fisherfolk, the Court simply stated that it was ‘not convinced that the provisional equidistance line would entail such harsh consequences for the population of Kenya’,[19] without fully engaging with what these ‘harsh consequences’ might refer to.
Problems with the Court’s current approach
Some scholars have accepted that Somalia v Kenya is consistent with case law.[20] For instance, Ionnades, Yiallourides, and Lath argue that the Court’s simplistic reading of non-geographical factors is unproblematic because its focus on geographical criteria, such as concavity and coastal length, aligns with delimitation jurisprudence.[21] Some may also argue that the significance of non-geographical factors is often not used to amend a de novo (without reference to the Court’s previous legal conclusions) provisional line, but to justify the Court’s conclusion that a maritime boundary already existed between the Parties.[22]
However, this is precisely the most problematic aspect of the Court’s jurisprudence – it does not go far enough in engaging with the role of non-geographical factors in maritime delimitation. In Jan Mayen, the Court was tasked to delimitate Denmark and Norway’s fishing zones and continental shelf areas in the waters between the east coast of Greenland and the Norwegian island of Jan Mayen. The Court considered whether the ‘size and special character of Jan Mayen’ and ‘absence of locally-based fishing’ were important circumstances affecting the provisional boundary. However, the Court eventually adopted a mechanical view of sovereignty, concluding that the legal rules governing maritime delimitation are based solely on the extent of permanent territory a State possesses, and do not ‘leave room for any considerations of economic development’.[23] This strict standard for considering non-geographical economic factors was established despite the Court’s initial proclamation that it had to consider whether amending the provisional line was required ‘to ensure equitable access to the capelin fishery resources for the vulnerable fishing communities concerned’.[24]
Moreover, in Somalia v Kenya, the Court did not adequately explain why the non-geographical factors proposed by Kenya failed to meet the high standards for acceptance set in previous cases. This creates a double-problem whereby, firstly, there is an existing gap between the Court’s commitment to equitable amendment and its exacting threshold, and, secondly, the strictness of this threshold is itself taken as reason not to meet it. For example, in concluding that the Jan Mayen standard was not reached, the Court simply referred to a single piece of evidence pointing to seventeen out of nineteen fish landing sites located near or at the Lamu Archipelago, and ‘only two landing sites… close to the land boundary terminus’.[25] The Court further recalled its statement in Nicaragua v Colombia where it established that legitimate security interests would only be relevant insofar as the maritime delimitation was constructed ‘particularly near to the coast of a State’, a threshold that was apparently not met here.[26] However, the lack of elaboration on what constitutes ‘nearness to the coast’ is particularly jarring given that in Nicaragua the delimitation of the maritime area was 200 miles from the coasts of both Parties, after first concluding that the agreed maritime boundary extended to a distance of 80 nautical miles along the parallel from its starting point, which does not seem exceptionally ‘near’.[27] Hence, by simply stating that the ‘provisional equidistance line does not pass near the coast of Kenya’,[28] the Court failed to clarify the threshold where a maritime boundary would trigger security interests.
Moreover, despite the maxim that ‘the land dominates the sea’,[29] the Court in Somalia v Kenya paid surprisingly little attention to territorial disputes where non-geographical factors have been acknowledged. This is disappointing as the Court had previously stressed the importance of such factors in shaping states’ exercise of sovereignty. In Frontier Dispute (Burkina Faso v Niger), the Court expressed its wish that each Party should ‘have due regard to the needs of the populations concerned, in particular those of the nomadic or semi-nomadic populations’.[30] Moreover, it was asserted in Judge Trindade’s Separate Opinion that contemporary international law requires people and territory to be evaluated as a whole, rather than abstractions of the former being a subsidiary of the latter.[31] One might, indeed, argue that these statements refer more to states’ obligations rather than the Court’s responsibility to draw maritime boundaries. However, it is counterproductive to suggest that within these boundaries, states must account for the needs of local populations, but that this consideration can be blatantly disregarded in the actual shaping of the boundaries. This is because international law applies equal, if not stricter, principles to boundary formation than internal governance. For example, the use of force is prohibited as a means of acquiring territory or establishing a state under Article 2(4) of the UN Charter,[32] while armed attacks within a state’s internal boundaries would only be illegal if it reaches the threshold of more serious acts such as crimes against humanity.[33] Hence, in maritime delimitation cases, if states are required to consider the needs of local populations within their boundaries, this should equally be considered a factor in the delineation of these boundaries.
Thus, it is clear that while the Court did not fully ignore the existence of non-geographical factors in Somalia v Kenya, it did not go far enough in convincingly justifying why it declined to accept them. This mirrors a historical pattern of glossing over non-geographical considerations despite separately affirming their importance. Moreover, this creates a double standard between land delimitation and maritime delimitation cases, contradicting the interdependence of territorial and maritime sovereignty in international law.
Strict adherence to judicial formalism: the Court’s misinterpretation of its own judicial role
On a deeper level, the Court’s fidelity to geographical factors in Somalia v Kenya is doctrinally unsound because it binds itself to a formalist interpretation of sovereignty, without appreciating how these maritime areas are used. This is problematic because the Court approaches sovereignty as an immutable, stable, and unchanging concept to be ascertained, a highly Eurocentric approach which ignores topophilic considerations of human geography and movement.[34]
Weil argues that the reason why geographical considerations are so entrenched in maritime delimitation is because the very philosophy of maritime jurisdiction requires it. He suggests that maritime rights are not autonomous rights but extensions of territorial rights over land. As such, the location of the delimitation line is contingent upon the coastal configuration of a state’s territory.[35] Additionally, non-geographical factors are used as a rhetorical tool to reinforce other arguments because the Court prioritises the principle of stability of boundaries and states prioritise their territorial integrity.[36] The primacy of territorial sovereignty (authority to exercise control over defined territory), of which maritime entitlement (rights to areas of the sea) is merely a subset, is evinced by how courts often treat the former as the overarching issue when both are in dispute.[37] Hence, this explains why the overarching legal framework on maritime jurisdiction is inherently geographical in nature – the main legal instrument, the United Nations Convention on the Law of the Sea (‘UNCLOS’) hinges on geographical, geophysical and hydrographic terminology,[38] such as by referring to distances, coordinates, and straight lines, and adhering to the idea that ‘the land dominates the sea’.[39]
In Gulf of Maine, the Court further argued that taking into account non-geographical considerations would amount to a decision ex aequo et bono.[40][41] Thus, it conceptualised the ‘relevant considerations’ stage as a strict application of geographical calculations followed by bare regard to non-geographical factors simply in relation to geographical considerations.
However, by construing geographical considerations as ‘on the basis of law’ and relegating non-geographical considerations outside the legal framework,[42] the Court ignores the gap between the formalistic application of legal rules and lived realities of how this territory is used. The very idea of sovereignty and self-determination[43] in international law, developed under the nexus of geographic emphasis and judicial formalism, negates the social realities of populations in these emerging states.[44] Essentially, the Court views its adjudicative role as one based on the plain meaning of legal rules, rather than a simultaneous application of historical, purposive, and functional interpretative premises.[45] This derives from an approach frequently invoked in frontier disputes whereby formalism is lauded as promoting a rational, deductive, and predictable application of the law.[46]
However, there is a gap between the illusory simplicity of the judicial role to apply and communicate legal rules regarding boundary delimitation, and the extensive network of rights and claims that a state derives from these boundaries, at the expense of its neighbours. The law cannot claim to be simply subservient to a false idea of territorial stability while creating and restricting these rights as a result of its adjudication. In fact, the very principle of equity in maritime delimitation shows that legal formalism is not, and should not be, absolute.[47] While applied in a different context, Davies critiques the fiction that the law is an ‘autopoietic’ system, or a formal and technical body of rules independent of the lived experiences of its subjects. In other words, legal rules are not a self-organising system based on mechanical demarcations of sovereignty and control, but subject to the actions of heteronomous agents who interpret and are confined by these rules.[48][49] Thus, in excluding considerations like the relative economic position of two countries and ‘cultural factors’ such as the size and special character of a relevant population, the Court binds itself to a formalistic, technical approach of maritime delimitation which artificially rejects how these maritime zones are used and inhabited. This is highly problematic because, while the Court has often been seen as a mechanism to transcend or regulate socio-political realities, it has historically ruled on cases involving major geopolitical tensions. Hence, excessive deference to judicial formalism would simply negate the mutually constitutive relationship between international law and social relations.[50]
Additionally, it is fallacious to presuppose that deference to geographical considerations creates greater predictability in delimiting maritime zones. While geographical factors are seen as more objective by the Court,[51] they are still subject to judicial discretion.[52] For instance, in the Black Sea case, the Court eliminated ex proprio motu[53] the seaward end of the Sulina dyke and any points on Snake Island, simply due to the inconsequential size of the latter, contradicting common practice where the provisional line would be strictly drawn.[54] Moreover, in the 2014 Bay of Bengal Maritime Arbitration, the Tribunal inflated the length of the relevant coastline of India by including a part of the Andaman Islands in its assessment, which was a substantial advantage conferred without India’s request. Both cases illustrate that geographical factors are still subject to judicial latitude rather than providing any precise, immutable geophysical truth. Thus, in stressing that a key reason for its deference to the ‘three-stage’ test was that it ‘brought predictability to the process of maritime delimitation’,[55] the Court should not have implicitly equated this predictability to purely geographical factors.
Conclusion and Recommendations
Overall, this piece has argued that the Court should move beyond a narrow reading of ‘relevant circumstances’ as purely geographical, an interpretation it continues to be confined by in Somalia v Kenya. As argued above, an overly formalist understanding of sovereignty predicated on the stability of boundaries is misplaced. Thus, the Court should consider non-geographical factors with equal weight at the same stage of the test, such that ‘relevant circumstances’ encompasses both geographical factors and other factors such as local populations, economies, and national security. The specific non-geographical considerations would differ based on the case and parties involved, but they should be addressed in depth rather than tangentially. A ‘checklist’ of various non-geographical considerations could also be useful, allowing the Court to methodically assess whether it has sufficiently evaluated various aspects such as the welfare and livelihood of local populations. If implemented, this checklist should not be static but flexible and evolutionary according to non-geographical factors that may gain greater weight over time, such as the impact of a state’s practices on climate change in the relevant maritime area. Ultimately, this proposal aims to remove maritime adjudication from an overly formalist view of boundary delimitation and excessively superficial consideration of non-geographical factors, especially given the plurality and extensiveness of rights that maritime boundaries confer upon states. In short, a more meaningful application of the three-stage test can only be achieved if the Court incorporates non-geographical considerations with equal scrutiny.
[1] The earliest precedent is that of Maritime Delimitation in the Black Sea (Romania v Ukraine) [2009] ICJ Rep 61, ICGJ 6.
[2] United Nations Convention on the Law of the Sea (adopted 10 December 1982, entered into force 16 November 1994) 1833 UNTS 3 (‘UN Convention on the Law of the Sea’) arts 3, 5, 7, 33, and 57.
[3] SP Jagota, Maritime Boundary (Martinus Nijhoff 1985).
[4] UN Convention on the Law of the Sea (n 2) art 56.
[5] John CP Goldberg and Benjamin C Zipursky, Hohfeldian Analysis and the Separation of Rights and Powers (Cambridge University Press 2022).
[6] Somalia v Kenya [2021] ICGJ 555 [26]. Essentially, Somalia asked the Court to fix the maritime boundary according to the equidistance principles set out in Articles 3, 5, 7, 33 and 57, without further modification based on ‘relevant circumstances’.
[7] ibid.
[8] Somalia v Kenya [2017] ICGJ 508 [35].
[9] Somalia v Kenya (n 6) [142].
[10] ibid [149].
[11] Maritime Delimitation in the Indian Ocean (Somalia v Kenya) (Preliminary Objections of the Republic of Kenya, 2015) <Somalia v Kenya. Preliminary Objection, October 2015. FINAL .docx> accessed 9 March 2025 [16].
[12] Somalia v Kenya (n 6) [152].
[13] Maritime Delimitation in the Indian Ocean (Somalia v Kenya) (Reply of Somalia, 2018) <FHE-DC-#186249-v4-REPLY_OF_SOMALIA> accessed 9 March 2025 [2.68(b)]-[2.69].
[14] ibid [2.70].
[15] Maritime Delimitation in the Indian Ocean (Somalia v Kenya) (Rejoinder of the Republic of Kenya, 2018) <1.1. TITLE PAGE VOL 1.docx> accessed 9 March 2025 [89].
[16] Somalia v Kenya (n 6) [153].
[17] Somalia v Kenya (n 8) [124].
[18] Somalia v Kenya (n 6) [158].
[19] ibid [159].
[20] Clive Schofield, Pieter Bekker, and Robert van de Poll, ‘The World Court Fixes the Somalia-Kenya Maritime Boundary: Technical Considerations and Legal Consequences’ (2021) 25(25) American Society of International Law Insights 1.
[21] Nicholas A Ionnades and Constantinos Yiallourides, ‘A Commentary on the Dispute Concerning the Maritime Delimitation in the Indian Ocean’ (Blog of the European Journal of International Law, 22 October 2021) <https://www.ejiltalk.org/a-commentary-on-the-dispute-concerning-the-maritime-delimitation-in-the-indian-ocean-somalia-v-kenya/> accessed 21 August 2024; Adhiraj Lath, ‘Taming Maritime Boundaries: The ICJ’s Decision in Somalia v Kenya’ (Jindal Forum for International and Economic Laws, 14 July 2022) <Taming Maritime Boundaries: The ICJ’s Decision in Somalia v Kenya – Jindal Forum for International and Economic Laws> accessed 9 March 2025.
[22] An example would be Peru v Chile [2014] ICJ Rep 4, ICGJ 473(refer, in particular, to [100]-[102]), where, in establishing a maritime boundary falling below the 200-nautical-mile limit, the Court treated the impact on fishing populations as evidentiary rather than determinative. Moreover, the Court did not use the ‘three-stage’ approach, instead relying on the sources of international law to find the existence of a tacit agreement between the Parties.
[23] Jan Mayen (Denmark v Norway) [1993] ICJ Rep 38, ICGJ 94 [80].
[24] ibid [75].
[25] Somalia v Kenya (n 6) [159].
[26] ibid [158].
[27] Territorial and Maritime Dispute (Nicaragua v Colombia) [2012] ICGJ 436 [152]-[176].
[28] Somalia v Kenya (n 6) [158].
[29] North Sea Continental Shelf Cases (Federal Republic of Germany/Denmark; Federal Republic of Germany/Netherlands) [1969] ICJ Rep 3 [96].
[30] Frontier Dispute (Burkina Faso v Niger) [2013] ICGJ 491 [112].
[31] ibid, Separate Opinion of Judge Trindade [62].
[32] Charter of the United Nations (adopted 26 June 1945, entered into force 24 October 1945) 1 UNTS XVI, art 2(4).
[33] Rome Statute of the International Criminal Court (adopted 17 July 1998, entered into force 1 July 2002) 2187 UNTS 90, art 7.
[34] Christopher R Rossi, ‘The Widening Gyre: Legal Formalism and International Law’s Sense of Place’ (2021) 11(1) Notre Dame Journal of International and Comparative Law 112, 131-135.
[35] Prosper Weil, ‘Geographic Considerations in Maritime Delimitation’ in Jonathan I Charney and Lewis M Alexander (eds), International Maritime Boundaries (Dordrecht, Boston, London, Martinus Nijhoff Publishers 1993) 115.
[36] Yusra Suedi, ‘Self-determination in Territorial Disputes before the International Court of Justice: From Rhetoric to Reality?’ (2022) 36(1) Leiden Journal of International Law 1. Suedi’s argument centres around self-determination rather than non-geographical factors, but because self-determination relates to the legal right of people to pursue their economic, social and cultural development, the argument can equally be applied to non-geographical factors because both require a non-territorial, geographical concept of sovereignty.
[37] Pierre-Emmanuel Dupont, ‘Contested Sovereignty over Land Territory and Maritime Zones’ in Stephen Minas and H Jordan Diamond (eds), Stress Testing the Law of the Sea (Martinus Nijhoff 2018).
[38] Clive Schofield, ‘Geographical Dimensions to Global Oceans Governance’ (2021) 113(1) Geographical Review 20.
[39] Territorial and Maritime Dispute (Nicaragua v Colombia) [2012] ICGJ 436 [140].
[40] Translation: ‘on the basis of what’s fair and good’.
[41] Delimitation of the Maritime Boundary in the Gulf of Maine Area (Canada/United States of America) [1984] ICJ Rep 246, ICGJ 121 [59].
[42] ibid.
[43] Refer to art 1(2) of the UN Charter and art 1 of the International Covenant on Economic, Social and Cultural Rights (adopted 16 December 1996, entered into force 3 January 1976) 993 UNTS 3 (ICESCR).
[44] Steven R Ratner, ‘Drawing a Better Line: Uti Possidetis and the Borders of New States’ (1996) 90 American Journal of International Law 590.
[45] Marcin Matczak, ’Why Judicial Formalism is Incompatible with the Rule of Law’ (2018) 31 Canadian Journal of Law and Jurisprudence 61.
[46] Christopher R Rossi, ‘The Widening Gyre: Legal Formalism and International Law’s Sense of Place’ (2021) 11(1) Notre Dame Journal of International and Comparative Law 107.
[47] Yusra Suedi, ‘Man, Land and Sea: Local Populations in Territorial and Maritime Disputes before the International Court of Justice’ (2021) 20 The Law and Practice of International Courts and Tribunals 30, 44.
[48] Margaret Davies, ‘EcoLaw: Legality, Life, and the Normativity of Nature’ (Routledge 2022).
[49] Continental Shelf (Libyan Arab Jamahiriya v Malta) [1985] ICJ Rep 13, ICGJ 118 [50].
[50] Gleider Hernández, ’High Politics and the International Court of Justice’ (National University of Singapore CIL Dialogues, 17 October 2024) <https://cil.nus.edu.sg/blogs/high-politics-and-the-international-court-of-justice/> accessed 3 February 2025.
[51] Somalia v Kenya(n 6) [128].
[52] Malcolm D Evans, ‘Maritime Boundary Delimitation: Where Do We Go From Here?’ in Richard Barnes, David Freestone, and David Ong (eds), The Law of the Sea: Progress and Prospects (Oxford University Press 2015) 254.
[53] This refers to an action or decision taken independently by the Court, without request by any of the Parties involved.
[54] Colter G Lathrop, ‘Maritime Delimitation in the Black Sea (Romania v Ukraine)’ (2009) 103 American Journal of International Law 543.
[55] Somalia v Kenya (n 6) [128].
Gillian Choy
LLB (LSE) ’25 and Notes Editor of the LSE Law Review Summer Board 2023-24. The author would like to thank Dr Lora Izvorova (LSE) and the LSE Law Review Editorial Team for their valuable assistance.